iBMQ

iBMQ performs integrated hierarchical Bayesian modeling to map expression quantitative trait loci (eQTLs) by jointly analyzing single nucleotide polymorphisms (SNPs) and gene expression levels to elucidate gene regulatory architecture.

Key Features:

• Hierarchical Bayesian Framework: Employs an integrated hierarchical Bayesian model that jointly analyzes all genes and single nucleotide polymorphisms (SNPs).
• Handling High Dimensionality: Tailored for the "large G, large S, small n" paradigm with many gene expressions (G), many genetic markers or regressors (S), and relatively small sample size (n).
• Strength Borrowing: Borrows strength across all gene expression data to improve mapping accuracy and robustness.
• False Positive Control: Incorporates mechanisms to control the number of false positives in eQTL detection.

Scientific Applications:

• eQTL mapping and gene regulation: Maps associations between SNPs and gene expression to elucidate gene regulatory architecture underlying complex traits.
• eQTL hotspot identification: Identifies eQTL hotspots, regions with high density of significant SNP–expression associations.
• Annotation enrichment and pathway inference: Reveals enrichment of genes within specific annotation categories to inform biological pathways and regulatory networks.

Methodology:

Validation using simulation studies across various scenarios and application to real expression datasets from BXD Recombinant Inbred (RI) mouse strains, with performance compared to QTLBIM, R-QTL, remMap, and M-SPLS.